3 ms·
Great suggestions. I'll probably structure the course in the stats/ml vein--that corresponds to what I already had in mind. Do you have any recommendations for
by tomrod 13y ago
Great suggestions. I'll probably structure the course in the stats/ml vein--that corresponds to what I already had in mind.
Do you have any recommendations for big data components? Would it be worth teaching how to use a Hadoop cluster, or is a small toy cluster too far abstract from what using a large cluster requires?
- eshvk 13y agoWell, I took a couple of courses at UT which dealt with "practical" distributed systems. The first was more on the lines of here is a large dataset, how do we design a distributed system that handles it. There, we learn about principles/paradigms such as MPI, threading, CUDA to address these issues. I took another course that was basically only Hadoop. Both are good. It completely depends on what the outcome of your course is: If you are looking for someone who just needs a flavor of what it means to think at that scale, generating an EC2 cluster and working on a few Map Reduce problems should be fine. If you are looking more to inculcate general principles of how to think at scale, the former should be good.